European Journal of Operational Research 226 (2013) 221227

Contents lists available at SciVerse ScienceDirect

European Journal of Operational Research

journal homepage: www.elsevier.com/locate/ejor

Production, Manufacturing and Logistics

Dual-channel closed-loop supply chain with government consumption-subsidy

Wei-min Ma, Zhang Zhao , Hua Ke
School of Economics and Management, Tongji University, Shanghai 200092, China

a r t i c l e i n f o a b s t r a c t

Article history: The government has been acting as an important role in the formation and operation of closed-loop sup-
Received 29 September 2011 ply chain. This paper focuses on how consumption-subsidy inuences dual-channel closed-loop supply
Accepted 29 October 2012 chain. After introducing government consumption-subsidy program and dual-channel closed-loop supply
Available online 7 November 2012
chain, the paper analyzes the channel members decisions before and after the government-funded pro-
gram performance, respectively. Finally, inuence of consumption-subsidy has been considered from the
Keywords: consumers, the scale of closed-loop supply chain and the enterprises perspectives, which provides an
Supply chain management
important basis for our propositions. The key propositions of the paper are listed as follows: All the con-
Consumption-subsidy
E-commerce
sumers that purchase the new products are beneciaries of the government consumption-subsidy in
China varying degrees; the consumption-subsidy is conducive to the expansion of closed-loop supply chain;
both the manufacturer and the retailer are beneciaries of the consumption-subsidy, while the e-tailer
benets or not is uncertain.
2012 Elsevier B.V. All rights reserved.

1. Introduction November 26, 2007, the Chinese government announced a

In recent years, the economic and environmental benets of
recovering obsolete products have been widely recognized in liter-
ature and practice. Closed-loop supply chain can generate prots
by taking back products from consumers and recovering
the remaining added value (Guide and Van Wassenhove, 2001;
Savaskan et al., 2004; Ferguson and Toktay, 2006; Geyer et al.,
2007; Schluep et al., 2009). Thierry et al. (1995) offered a system
model of closed-loop supply chain, and divided the product recov-
ery into ve options: repair, refurbishing, remanufacturing, canni-
balization, and recycling. Depending on the economics of a
particular situation, recovery process may reuse the entire product,
selected modules, components, and/or parts (Atasu et al., 2008).
Guide and Van Wassenhove (2009) introduced the eld of
closed-loop supply chains with a strong business perspective.
However, obsolete products (especially electronic wastes) can
cause environmental damage if not dealt with in appropriate ways
(Chi et al., 2011). Atasu et al. (2009) discussed the economic and
environmental impacts of extended producer responsibility type
of legislation and identied efciency conditions.
In China, the government has been acting as an important role
in the formation and operation of closed-loop supply chain. In or-
der to stimulate domestic consumption, curb pollution and devel-
op circular economy, the government announced detailed subsidy
plans for home appliance consumption and replacement: (1) on
Corresponding author. Tel.: +86 182 8861 9260; fax: +86 21 6598 6304.
E-mail address: 091zhaozhang@tongji.edu.cn (Z. Zhao).
government-funded program aiming to expand home appliance
sales in rural areas. According to the policy, rural consumers would
receive a consumption-subsidy worth 13% of the new appliances
price (Ministry of Commerce of the Peoples Republic of China,
2007) and (2) on June 1, 2009, the Chinese government announced
the detailed subsidy plan for home appliance replacements.
According to the policy, consumers will receive a replacement-sub-
sidy worth 10% of the price of ve kinds of new appliancestelevi-
sions, refrigerators, washing machines, air-conditioners and
computers (Chinas State Council General Ofce, 2009).
In this paper, we focus on how the consumption-subsidy inu-
ences the dual-channel closed-loop supply chain.
The internet has provided consumers a new way to interact
with their supply chain. Electronic commerce has been widely
hailed as a revolution to permanently transform the landscape of
consumer/supplier relationships. For example, according to a re-
search by China e-Business Research Center, Chinas fast-growing
e-commerce transactions hit 4.5 trillion yuan ($682.16 billion) in
2010, up 22% year-on-year. Of the 4.5 trillion yuan, online B2B
(Business-to-Business) sector accounted for 3.8 trillion yuan
($576.05 billion), increasing by 15.8% year-on-year. The online re-
tail business hit 513.1 billion yuan ($77.78 billion), a boost of 97.3%
from the previous year (China e-Business Research Center, 2011).
Additionally, during the nascent period of electronic commerce,
independent e-tailers played a dominant role in the development
of internet-based sales to consumers. For many product categories,
independent e-tailers continue to play a vital role in their
industrys supply chains. Hence, in this paper, we focus our
attention on the dual-channel closed-loop supply chain, in which

0377-2217/$ - see front matter 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ejor.2012.10.033
222 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227
a manufacturer, a traditional retailer and an internet-based retailer
each exist as independent entities.
This type of dual-channel with a traditional channel and an
internet-based channel have been widely studied. Chiang et al.
(2003) constructed a price-setting game between a manufacturer
and its independent retailer and showed that the mere threat of
introducing the direct channel can increase the manufacturers
negotiated share of cooperative prots even if price efciency is
obtained by using other business practices. Cattani et al. (2004) fo-
cused on the coordination opportunities that arise when a rm
participates in both a traditional channel and an internet channel.
Chiang and Monahan (2005) indicated that the dual-channel strat-
egy outperforms the other two channel strategies in most cases,
and the cost reductions realized by the exibility of the dual-
channel system may be signicant under some circumstances.
Cattani et al. (2006) analyzed a scenario where a manufacturer
with a traditional channel partner (i.e., a retailer) opens up a direct
Internet channel that is in competition with the traditional channel
partner.
Although all members decisions and closed-loop supply chains
indexes will be inuenced by the subsidy, to simplify the analysis
and to highlight the focus of our study, we only consider the man-
ufacturer, the retailer and the e-tailers decisions, and discuss the
subsidy inuence from the consumers, the scale of closed-loop
supply chain and the enterprises perspectives, respectively.
The remainder of the paper is organized as follows. The follow-
ing section is devoted to the model conceptualization and formula-
tion. Section 3 analyzes the channel decisions of closed-loop
supply chain models before and after the government-funded pro-
gram performance, respectively. Through comparing the corre-
sponding results obtained in Section 3, we explain how the
consumption-subsidy inuences the dual-channel closed-loop
supply chain in Section 4. Finally, conclusions are provided in the
last section.
2. Model assumptions and notations
The goal of this paper is to develop an understanding of the
inuence of consumption-subsidy on the consumers, the scale of
the closed-loop supply chain and the enterprises. To this end, we
formulate a basic dual-channel closed-loop supply chain model
(Model N, Fig. 1a), in which new products are produced by the
manufacturer and sold by the retailer and the e-tailer. Based on
this model, we formulate and analyze another closed-loop supply
chain model (Model S, Fig. 1b), in which the consumers can receive
consumption-subsidy.
In the rest of the paper, the following notations dened in
Table 1 are used.
For the sake of simplicity, we consider the following scenario
and make the following modeling assumptions.
Assumption 1. We consider closed-loop supply chains with one
single kind of product. A family should own one and only one
product, whether new or obsolete.
(a) Model N
Primary
Manufacturer Retailer consumers
Forward
supply chain E-tailer Replacement
consumers
Reverse supply chain
Fig. 1. Closed-loop supply chain models.
Table 1
Notations.
Parameter Denition
cm Cost of manufacturing a new product
w Wholesale price of a new product charged by the manufacturer
cr Cost of selling a new product by retailer
ce Cost of selling a new product by e-tailer
pr Price of a new product charged by the retailer
pe Price of a new product charged by the e-tailer
po Price of an obsolete product
D Market size
b Ratio of the replacement consumer
hP Willing to pay of the primary consumer for the new product sold
by the retailer
hR Willing to pay of the replacement consumer for the new product
sold by the retailer
d Ratio between the willing to pays for the products sold by the e-
tailer and the retailer
u Subsidy ratio
qki Quantity of the new products sold by channel member i and
purchased by consumer k. Superscript k takes the values of P, R,
denoting the primary and replacement consumers, respectively.
Subscript i takes the values of r and e, denoting the retailer and
the e-tailer, respectively. (Additionally,
qP qPr qPe ; qR qRr qRe ; qr qPr qRr ; qe qPe qRe and
q = qP + qR = qr + qe. Certainly, the quantity of the obsolete
product recycling is equal to qr)
Pki Prot function for channel member i in the Model k. Superscript
k takes the values of N, S, denoting Models N and S, respectively.
Subscript i takes the values of m, r and e, denoting the
manufacturer, the retailer and the e-tailer, respectively
According to the policy in China, the government subsidizes the
consumption of many kinds of home appliances, and there are dif-
ferent brands of each kind of home appliance. But for simplicity,
we consider the closed-loop supply chain with a single kind of
product. The last part of Assumption 1 tallies with the actual
situation.
Assumption 2. The obsolete products are identical, and only used
to be recycled.
Because of the rapid development of science and technology,
home appliances are very easily outdated. In the real situation,
most of obsolete home appliances can only be used for recycling,
and their values are similar. So for simplicity, we consider the ob-
solete products as identical.
Assumption 3. While optimizing their objective functions, all
closed-loop supply chain members have access to the same
information. The manufacturer has sufcient channel power over
the retailer and the e-tailer to act as a Stackelberg leader.
The former part of this assumption enables us to control inef-
ciencies and risk-sharing issues resulting from information asym-
metry. The similar forms of this assumption have been widely
used (Savaskan et al., 2004). The latter of this assumption states
that the manufacturer uses his/her foresight about the retailer
and the e-tailers reaction functions in her decision making. The
(b) Model S
Primary
Government
Manufacturer Retailer consumers
Forward Replacement
supply chain E-tailer
consumers
Reverse supply chain
 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227 223

Stackelberg structure for the solution of similar games has been and U Pe > 0. For the replacement consumers, when
widely used in the supply chain literature (Savaskan et al., 2004; h 1; U Rr > U Re and U Rr > 0; when hR pde ; U Re > U Rr and U Re > 0.
R

Tayur et al., 1998). Thus, pper < d < 1  pr pe .

Similarly, In Model S, U Pr  U Pe 1  dhP  1  upr  pe and
Assumption 4. There are two types of consumers: primary U Rr  U Re 1  dhR  1  upr  pe . In order to ensure the exis-
consumers (P) and replacement consumers (R). The primary tence of dual-channel closed-loop supply chain, the consumers
consumer does not have an obsolete product and can purchase a whose wtp is 1 must purchase new product from the retailer,
new product directly. The replacement consumer has an obsolete and the consumers whose wtp is 1dupe must purchase new product
product and should sell his/her obsolete one when he/she
from the e-tailer. Thus, pper < d < 1  1  upr  pe .
purchases a new one. During a certain period, the market size is
D, and the ratio of the replacement consumer is b. Because 1  pr + pe < 1  (1  u)(pr  pe), we assume
pe
pr
< d < 1  pr pe .

Assumption 5. The primary and replacement consumers are het- Assumption 8. The closed-loop supply chain decisions are con-
erogeneous with hP and hR, respectively. And hP and hR are assumed sidered in a single-period setting.
to be uniformly distributed between 0 and 1. We assume the previous existence of the product in the market.
The similar forms of the last part of this assumption have been Those products sold in the previous periods can be recycled. The
widely used in the consumer segmentation (Chiang and Monahan, similar forms of this assumption have been widely used (Savaskan
2005; Debo et al., 2005; Atasu et al., 2009). On the one hand, the et al., 2004).
consumers are willing to pay more for the retailers than the e-tai-
ler products. On the other hand, the cost of selling a new product
3. Model formulation and solution
by retailer is more than the one by e-tailer (cr > ce). These are the
fundamental reasons why the dual-channel can be formed.
This section presents the manufacturer, the retailer and the e-
tailers decisions in different models, respectively.
Assumption 6. The consumption-subsidy is related to the new
Before analyzing the game between the channel members, we
products price. When purchasing the new product from the
must discuss the demand functions in the different models.
retailer, the consumer can get subsidy upr, 0 < u < 1. And when
According to the assumptions in the last section, we may conclude
purchasing the new product from the e-tailer, the consumer can
the following:
get subsidy u pe.
This assumption is merely for simplicity and comparison. (1) In Model N, for the primary consumers whose wtp is in the
From the above assumptions, we draw the following  r pe  P
interval p1d ; 1 , U r > U Pe and U Pr > 0; for the replacement
conclusions:  r pe  R
  consumers whose wtp is in the interval p1d ; 1 , U r > U Re
(1) In Model N, a primary consumer gets utility U Pr hP  pr and U Rr > 0; For the primary consumers whose wtp is in
 
and U Pe dhP  pe via purchasing a new product from   P
the interval pde ; p1d r pe
, U e > U Pr and U Pe > 0; For the replace-
the retailer and the e-tailer, respectively. Similarly, a  
  ment consumers whose wtp is in the interval pe p d
o pr pe
; 1d ,
replacement consumer gets utility U Rr hR  pr po and  
  U Re > U Rr and U Re > 0. Thus, qPr D1  b 1  p1d r pe
, qPe
pr pe pe  R    
U Re dhR  pe po via purchasing a new product from pr pe
D1  b 1d  d , qr Db 1  1d , qe Db 1d  d ; R pr pe pe po

the retailer and the e-tailer, respectively; (2) In Model S, for the primary consumers whose wtp is in the
h i
(2) In Model S, a primary consumer gets utility interval 1u1d pr pe
; 1 , U Pr > U Pe and U Pr > 0; For the replace-
   
U Pr hP  pr upr and U Pe dhP  pe upe via pur- h i
ment consumers whose wtp is in the interval 1u1d pr pe
;1 ,
chasing a new product from the retailer and the e-tailer,
respectively. Similarly, a replacement consumer gets utility U Rr > U Re and U Rr > 0; for the primary consumers whose wtp
    h i
U Rr hR  pr po upr and U Re dhR  pe po upe is in the interval 1dupe ; 1u1d
pr pe
, U Pe > U Pr and U Pe > 0;
via purchasing a new product from the retailer and the e-tai- for the replacement consumers whose wtp is in the interval
h i
1upe po 1upr pe
ler, respectively. d
; 1d
, U Re > U Rr and U Re > 0. Thus, qPr
h i h i
The primary consumer purchases a new product from the retai- D1  b 1  1u1d pr pe
, qPe D1  b 1u1d pr pe
 1dupe ,
h i h i
ler if U Pr > U Pe and U Pr > 0, and purchases a new product from the e- qRr Db 1  1u1d pr pe
, qRe Db 1u1d
pr pe
 1up e po
.
d
tailer if U Pe > U Pr and U Pe > 0. The replacement consumer purchases
a new product from the retailer if U Rr > U Re and U Rr > 0, and pur-
3.1. Model N no subsidy
chases a new product from the e-tailer if U Re > U Rr and U Re > 0.

pe A benchmark for describing how the consumption-subsidy

Assumption 7. pr < d < 1  pr pe .
inuences the dual-channel closed-loop supply chain is provided
by Model N.
In Model N, U Pr  U Pe 1  dhP  pr pe and U Rr  U Re Because the manufacturer is the Stackelberg leader, we begin by
1  dhR  pr pe . The consumers with higher wtp (willingness characterizing the best-response functions of the retailer and the
to pay) perfect to purchase new product from the retailer. In order e-tailer. For a given w, the retailer and the e-tailers problems are
to ensure the existence of dual-channel closed-loop supply chain,  p  pe 
the consumers whose wtp is 1 must purchase new product from max PNr qr pr  w  cr D 1  r pr  w  cr ;
pr 1d
the retailer, and the consumers whose wtp is pde must purchase  
new product from the e-tailer. In other words, for the primary con- p  pe pe  bpo
max PNe qe pe  w  ce D r  pe  w  ce :
sumers, when hP 1; U Pr > U Pe and U Pr > 0; when hP pde ; U Pe > U Pr pe 1d d
224 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227

From the concavity of the objective functions, we determine the comparison, we tag the superscript 0 to the symbols under Model
best response functions from the rst-order conditions for pr and S, e.g. p0r denotes the unit price of a new product in the retailer un-
pe. Thus, one can easily show that der Model S.
1
pr 21  d 3w 2cr ce b1  dpo ; 4.1. Consumers perspectives
4d
1
pe d1  d 2 dw dcr 2ce 2b1  dpo : Through analyzing the assumptions and the results in the pre-
4d
ceding sections, we can generate the following results: In Model
 r pe 
Given the retailer and the e-tailers response functions, the N, the consumers whose wtp is in the interval p1d ; 1 purchase
manufacturer solves the new products from the retailer. The primary consumers whose
 
p  bpo p 
max PNm q w  cm D 1  e w  cm ; wtp is in the interval d
e
; p1d
r pe
and the replacement consumers
w d
p 
to determine the new products wholesale price. whose wtp is in the interval e po
d
; p1d
r pe
purchase the new products
Because the objective function is concave in w, one can easily
show that the optimal new products wholesale price is given by from the e-tailer; In Model S, the consumers whose wtp is in the

1up0r p0e
1 interval ; 1 purchase the new products from the retai-
w 3d 2 dcm  dcr  2ce b2 dpo : 1d
22 d
ler. The primary consumers whose wtp is in the interval
Optimal new products prices and equilibrium new product 
1up0e 1upr pe
0 0
sales, quantity of the obsolete product recycling and channel prof- ; and the replacement consumers whose wtp is
d 1d
its can easily be found by substitution of w. To improve readabil- 
1up0e po 1upr pe
0
ity, the results are listed in Table 2 in Appendix A. in the interval ; purchase the new products
d 1d

3.2. Model S consumption-subsidy from the e-tailer. Fig. 2 shows the retailer and the e-tailers sales
under different models.
In this model, the consumers will receive a unit subsidy when 1up0r p0e u
Because 1d
 p1d
r pe
21d2d4d 1  d2 dcm
he/she replaces his/her product.
1up0e 0
Just as in Model N, for a given w, the retailer and the e-tailer 8  d  d2 cr  6ce  < 0 and d
 pde 1up
d
e po
 pe p
d
o

solve u
 2d4d
dcm dcr 2ce  < 0,
2 we can draw the following
S 1  upr  pe propositions:
max P qr pr  w  cr D 1 
r pr  w  cr ;
pr 1d
  r pe 
1  upr  pe 1  upe  bpo (1) For the consumers whose wtp is in the interval p1d ; 1 , sub-
max PSe qe pe  w  ce D 
pe 1d d sidies do not inuence their choices. They purchase the new
pe  w  ce : products from the retailer. And due to the governments sub-


Because the objective functions are concave in pr and pe, respec- sidy program, they will save pr  p0r  up0r . It is a xed
u2cr ce
tively, the retailer and the e-tailers rst-order conditions charac- value and equals to 4d ;
terize the unique best response, (2) For the consumers whose wtp is in the interval

 1up0r p0e p p
1 21  d b1  d 1d
; r
1d
e
, subsidies inuence their choices. Due to
pr 3w 2cr ce po ;
4d 1u 1u the governments subsidy program, they will no longer

 1 d1  d 2b1  d purchase the new products from e-tailer, but to purchase
pe 2 dw dcr 2ce po :
4d 1u 1u the new productsfrom
the retailer.  Every primary consum-
ers will save hP  p0r  up0r  dhP  pe , and every
The manufacturers problem can be stated as: 
 
 replacement consumers will save hP  p0r  up0r po
P
1  upe  bpo dh  pe po ;
max PSm q w  cm D 1  w  cm :
w d (3) For the primary consumers whose wtp is in the interval

pe 1upr pe
0 0
Again, the objective function is concave in w, and the manufac- d
; 1d
and the replacement consumers whose wtp

turers rst-order condition characterizes the unique best 1up0r p0e
response, is in the interval pe p d
o
; 1d
, subsidies do not inu-
 ence their choices. They will purchase the new products
1 3d b2 d
w 2 dcm  dcr  2ce po : from the e-tailer. Due to the governments subsidy program,
22 d 1  u 1u
they will save pe  p0e  up0e . It is a xed value and equals to
Optimal new products prices and equilibrium new product udcr 2ce
4d
;
sales, quantity of the obsolete product recycling and channel prof- (4) For the primary consumers whose wtp is in the interval
its can easily be found by substitution of w. The results are also h i
1up0e pe
; d and the replacement consumers whose wtp is in
listed in Table 2 in Appendix A. d
h i
1up0e po pe po
the interval d
; d , subsidies inuence them obvi-
4. Inuence of consumption-subsidy ously. Without the governments subsidy program, they will
not purchase any product. And with the governments sub-
This section will focus on how consumption-subsidy inuences sidy program, they will purchase the new products from
the dual-channel closed-loop supply chain from the consumers, the e-tailer. The primary and the replacement consumers



the scale of closed-loop supply chain and the enterprises can get utility hP  p0e  up0e and hP  p0e  up0e po from
perspectives, and reach a series of propositions. For the sake of their purchases, respectively.
 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227 225

Fig. 2. The consumers choices and surpluses (D = 1, d = 0.9, u = 0.1, cm = 0.3, cr = 0.2, ce = 0.1, b = 0.5 and po = 0.05).

Both the primary consumers whose wtp is in the interval qP0 and qR0 . Thus, the subsidy augments the scale of the forward
h i supply chains.
1upe0
d
; 1 and the replacement consumers whose wtp is in the This subsection has clearly showed that the subsidy is condu-
h i
1up0e po cive to the expansion of closed-loop supply chain.
interval d
; 1 benet from the subsidy. And who gets the

most benet? The red lines in Fig. 2 give us a visual presentation. 4.3. Enterprises perspectives

This subsection focuses on how subsidy inuences the enter-

4.2. Scale of closed-loop supply chain
What is the scale of closed-loop supply chain? As we all know,
the closed-loop supply chain consists of forward and reverse sup-
ply chains. In this paper, we use the sale of new products and the
quantity of obsolete product recycling to represent the scales of the
forward and reverse supply chains, respectively.
From the assumptions and the analysis in the above, we know
that the replacement consumers whose wtp is in the intervals
h i h i
1upe po 1up0e po
d
; 1 and d
; 1 purchase new products from the re-
tailer or the e-tailer and sell their obsolete products to reverse sup-
ply chains under Model N and Model S, respectively. Through a
1up0e po
simple comparison, we can draw that 1up
d
e po
> d
. In other
words, qR0 > qR. It means that subsidy augments the scales of re-
verse supply chain. The quantity of the obsolete product recycling
equals to qR and qR0 under Model N and Model S, respectively. qR is
0 0
the sum of qRr and qRe . qR0 is the sum of qRr and qRe . In Fig. 2b, we will
nd the change easily.
Apart from replacement consumers, some primary consumers
also buy new products. The primary consumers whose wtp is in
h i h 0
i interval
the intervals 1dupe ; 1 and 1dupe ; 1 purchase the new products
from the retailer or the e-tailer under Model N and Model S,
respectively. Through a simple comparison, we can draw that
0
1upe
d
> 1dupe . In other words, qP0 > qP. qP is the sum of qPr and qPe .
P0 0 0
q is the sum of qPr and qPe . It means that more new products are
purchased by the primary consumers when the government subsi-
dizes the consumption. Fig. 2a shows this change clearly.
The sales of the new product are q and q0 under Model N and
Model S, respectively. q is the sum of qP and qR. q0 is the sum of
prises. To simplify the analysis, we just consider a basic dual-chan-
nel closed-loop supply chain, and only consider the decisions of the
manufacturer, the retailer and the e-tailer. The manufacturer
makes prot from manufacturing and wholesaling new products
to the retailer and the e-tailer, while the retailer and the e-tailer
make prot from selling new products to primary and replacement
consumers.
The manufacturers prot function (Pm = q(w  cm)) points out
that prot not only relates to total sales, but also relates to
wholesale price. From the above analysis, we know that subsidy
stumilates consumers to purchase more products. Additionally,
w0  w 21uu2d 3d b2 dpo  > 0. Thus, we can draw a con-
clusion that the subsidy increases the manufacturers prot.
Similarly, the retailers prot function (Pr = qr(pr  w  cr))
points out that prot not only relates to sales by himself, but also
relates to the new products price in retailer and wholesale price.
From the above analysis, we know that no matter there
 pis subsidy

or not, the consumers whose wtp is in the interval p1d r e
; 1 pur-
chase the new products from the retailer. And due to the govern-
ments subsidy program,
 the consumers whose wtp is in the
1up0r p0e p p
; 1d are no longer to purchase new products
r e
1d
from e-tailer, but to purchase new products from the retailer.
Therefore, subsidy stimulates consumers to purchase more prod-
ucts from the retailer. Additionally, though the wholesale price in-
creases after the government subsidizes consumption, new
products price in the retailer increases more as p0r  w0 
pr  w 21uu2d4d
1u
8 d b2 dpo  > 0. After the above
analysis, we can draw a proposition that the retailer is a beneciary
of the government consumption-subsidy.
226 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227

In this paper, we also consider the e-tailers decision, and how

1  d8 d  1  u1  d2 dcm  1  u8  d  d2 cr 61  uce b1  d2  dpo 2

d1  d2  1  u1  d2 dcm 3d1  ucr  1  u2 2d  d2 ce b1  d2 dpo 2

the consumption-subsidy inuences the e-tailer. The e-tailers
prot function (Pe = qe(pe  w  ce)) points out that prot not only
relates to sales by himself, but also relates to new products price in

b2 d5  2dpo 
e-tailer and wholesale price. From the above analysis, we know
that no matter there is subsidy or not, the consumers whose wtp

1up0r p0e
is in the interval pde ; 1d
purchase new products from the
e-tailer. The consumers whose wtp is in the interval

1up0r p0e p p
; r e
are no longer to purchase new products from e-

21u2d4d 8 5d  4d 31  2 dc m 1  8 dc r  21  1  dce
1d 1d

 d1  cr  21  ce 24  d  3b2  dpo g
tailer, but to purchase new products from the retailer. The primary

2 dcm d1  cr 21  ce 3b2  dpo 

h i

2
1up0e pe

4d1u2d4d 3d  1  2 dc m  d1  c r  21  c e b2  dpo 
consumers whose wtp is in the interval d
; d and the replace-
h i
1up0e po pe po
ment consumers whose wtp is in the interval d
; d pur-

d1  cr  21  ce b2  dpo 
chase new products from the e-tailer only after the government
subsidizes consumption. Fig. 2a shows this change clearly.

u
Whether sales by the e-tailer will increase or decrease is decided

u
1up0r p0e
by the relationship between p1d r pe
 1d
and pde 
 

u
u
1up0e 1 up 0 p
pe po
 e o
. Additionally, both wholesale price and

u
d d d

u
new products price increase after the government subsidizes
consumption. Which one increases more? It is also an uncertain

u
u
problem. Thus, it is an uncertain question that the e-tailer is

u
winner or loser of the government subsidy program.
This section has focused on how the consumption-subsidy


u
m

m
inuences the dual-channel closed-loop supply chain from the

dc
dc
u
2

1 
consumers, the scale of closed-loop supply chain and the enter-

2
prises perspectives, respectively. Through analyzing a series of

2d

41u1d2d2 4d2

d1u1d2d2 4d2
u
u
questions, we can come up with some propositions, such as




1
1
d5
changes of sales by the retailer and the e-tailer.




D
f3d
21u4d

3d
Model S

D
1

2d4d

2d4d
1

Db

D
5. Conclusions

For different purposes, various countries have issued series of

1  d8 d  1  d2 dcm  8  d  d2 cr 6ce b1  d2  dpo 2

measures attempting to inuence the closed-loop supply chain.

d1  d2  1  d2 dcm 3dcr  2 2d  d2 ce b1  d2 dpo 2

This paper focuses on the consumption-subsidy.
In the early phase of this research, we have introduced the gov-
b2 d5  2dpo 

ernment consumption-subsidy program and explained the reason

why we choose the dual-channel closed-loop supply chain, not
the single-channel one as the research object. Based on the intro-
duction and the explanation, we have made a number of assump-
tions to develop a more comprehensive understanding of different
dual-channel closed-loop supply chain models. Then, a series of
22d4d 8 5d  4d 32 dc m 8 dc r  21  dc e

dcr  2ce 24  d  3b2  dpo g

equilibrium channel indexes under different models have been

worked out. Finally, from the consumers, the scale of closed-loop
dcm dcr 2ce 3b2  dpo 

supply chain and the enterprises perspectives, we have considered

2
4d2d4d 3d  2 dc m  dc r  2c e b2  dpo 

the inuences of the consumption-subsidy, respectively.

r  2c e b2  dpo 
Comparison of different closed loop supply chain models.

The key propositions of the paper are listed as follows: all the
consumers that purchase the new products are beneciaries of
the government consumption-subsidy in varying degrees; the con-
sumption-subsidy is conducive to the expansion of closed-loop
supply chain; both the manufacturer and the retailer are benecia-
ries of the consumption-subsidy, while the e-tailer benets or not
dc


is uncertain.

2

m
dc
2

dc



2d

41d2d2 4d2
2

d1d2d2 4d2

Acknowledgments
2




f3d
d5

3d
Model N

The work was partly supported by the National Natural Science

2d4d

2d4d
D
1

24d
Db

Foundation of China (71071113, 71001080), a Ph.D. Programs

D
1

Foundation of Ministry of Education of China (20100072110011),

Shanghai Pujiang Program, and Shanghai Philosophical and Social
Table 2

Pm

Science Program (2010BZH003), the Fundamental Research Funds

Pe
Pr
qR
pe
pr

for the Central Universities.

 W.-m. Ma et al. / European Journal of Operational Research 226 (2013) 221227
Appendix A
See Table 2.
References
Atasu, A., Guide, V.D.R., Van Wassenhove, L.N., 2008. Product reuse economics in
closed-loop supply chain research. Production and Operations Management 17
(5), 483496.
Atasu, A., Van Wassenhove, L.N., Sarvary, M., 2009. Efcient take-back legislation.
Production and Operations Management 18 (3), 243258.
Cattani, K., Gilland, W., Heese, H.S., Swaminathan, J., 2006. Boiling frogs: pricing
strategies for a manufacturer adding a direct channel that competes with the
traditional channel. Production and Operations Management 15 (1), 4056.
Cattani, K.D., Gilland, W.G., Swaminathan, J.M., 2004. Coordinating traditional and
internet supply chains. In: Simchi-Levi, D. et al. (Eds.), Handbook of Quantitative
Supply Chain Analysis: Modeling in the E-Business Era. Kluwer, Boston, pp.
643677.
Chi, X.W., Streicher-Porte, M., Wang, M.Y.L., Reuter, M.A., 2011. Informal electronic
waste recycling: a sector review with special focus on china. Waste
Management 31 (4), 731742.
Chiang, W.K., Chhajed, D., Hess, J.D., 2003. Direct marketing, indirect prots: a
strategic analysis of dual-channel supply-chain design. Management Science 49
(1), 120.
Chiang, W.K., Monahan, G.E., 2005. Managing inventories in a two-echelon dual-
channel supply chain. European Journal of Operational Research 162 (2), 325
341.
China e-Business Research Center, 2011. 2010 China E-commerce Market Data
Monitoring Report. <http://b2b.toocle.com/zt/upload_data/down/2010dsjc.
doc> (accessed September, 2012, in Chinese).
227
Chinas State Council General Ofce, 2009. The State Council General Ofce
Document. <http://www.gov.cn/zwgk/2009-06/03/content_1331210.htm>
(accessed September, 2012, in Chinese).
Debo, L.G., Toktay, L.B., Van Wassenhove, L.N., 2005. Market segmentation and
product technology selection for remanufacturable products. Management
Science 51 (8), 11931205.
Ferguson, M.E., Toktay, L.B., 2006. The effect of competition on recovery strategies.
Production and Operations Management 15 (3), 351368.
Geyer, R., Van Wassenhove, L.N., Atasu, A., 2007. The economics of remanufacturing
under limited component durability and nite product life cycles. Management
Science 53 (1), 88100.
Guide, V.D.R., Van Wassenhove, L.N., 2001. Managing product returns for
remanufacturing. Production and Operations Management 10 (2), 142155.
Guide, V.D.R., Van Wassenhove, L.N., 2009. The evolution of closed-loop supply
chain research. Operations Research 57 (1), 1018.
Ministry of Commerce of the Peoples Republic of China, 2007. The Policies and
Regulations. <http://jdxx.zhs.mofcom.gov.cn/admin/news.do?method=view&id=
31879> (accessed September, 2012, in Chinese).
Savaskan, R.C., Bhattacharya, S., Van Wassenhove, L.N., 2004. Closed-loop supply
chain models with product remanufacturing. Management Science 50 (2), 239
252.
Schluep, M., Hagelueken, C., Kuehr, R., Magalini, F., Maurer, C., Meskers, C., Mueller,
E., Wang, F., 2009. Recycling from e-waste to resources, Final report. United
Nations Environment Programme and United Nations University, <http://
www.ewasteguide.info/les/UNEP_2009_eW2R.PDF> (accessed November,
2012).
Tayur, S., Ganeshan, R., Magazine, M., 1998. Quantitative Models for Supply Chain
Management. Kluwer Academic Publishers, Boston, MA.
Thierry, M., Salomon, M., Van Nunen, J., Van Wassenhove, L.N., 1995. Strategic
issues in product recovery management. California Management Review 37 (2),
114135.